摘要
为了减小桥式起重机箱型主梁自重,提出了基于动态随机重组粒子群算法的主梁尺寸优化方法。建立了箱型主梁截面积最小化目标函数,分析了主梁应力、强度、刚度、稳定性等约束,使用罚函数法将约束优化问题转化为了无约束优化问题。以粒子群算法为基础,依据粒子目标函数值将其分为近邻群、远亲群和重组群,并依据各子群的粒子特点设计了更新方法,有效平衡了粒子多样性和算法收敛性,并将新算法命名为动态随机重组粒子群算法,并将新算法应用于主梁尺寸优化。经主梁轻量优化验证可知,在满足约束前提下,动态随机重组粒子群算法优化的主梁面积均小于粒子群算法、文献[11]镜面反射算法、文献[12]GSA-GA算法,且性能参数在约束范围内,并留有较大性能裕度。实验数据和分析结果验证了动态随机重组粒子群算法在主梁优化中的优越性。
In order to reduce the dead weight of box girder of bridge crane,a girder size optimization method based on dynamic random reorganization particle swarm optimization algorithm is proposed.The objective function of minimizing the cross-sectional area of the box girder is established,and the constraints such as stress,strength,stiffness and stability of the main girder are analyzed.The constrained optimization problem is transformed into an unconstrained optimization problem by using the penalty function method.Based on particle swarm optimization algorithm,it is divided into nearest neighbor group,distant relative group and reorganization group according to the value of particle objective function.According to the particle characteristics of each subgroup,an update method is designed to effectively balance the particle diversity and algorithm convergence.The new algorithm is named dynamic random reorganization particle swarm optimization algorithm,and the new algorithm is applied to the size optimization of main beam.According to the verification of lightweight optimization of main beam,on the premise of meeting the constraints,the area of main beam optimized by dynamic random reorganization particle swarm optimization algorithm is smaller than that of particle swarm optimization algorithm,specular reflection algorithm in reference[11]and GSA-GA algorithm in reference[12],and the performance parameters are within the constraint range,leaving a large performance margin.The experimental data and analysis results verify the superiority of dynamic random reorganization particle swarm optimization algorithm in girder optimization.
作者
刘元华
郭乙运
李超群
LIU Yuanhua;GUO Yiyun;LI Chaoqun(Qingdao Huanghai University,International Business College,Shandong Qingdao 266427,China;Ocean University of China,Information Science and Engineering College,Shandong Qingdao 266100,China;Qingdao Port International Co.,Ltd.,Shandong Qingdao 266000,China;Qingdao Haina Electric Automation System Co.,Ltd.,Shandong Qingdao 266101,China)
出处
《机械设计与制造》
北大核心
2025年第12期103-107,112,共6页
Machinery Design & Manufacture
基金
国家自然科学基金(62072260)
青岛市自主创新重大专项(20322hy,211216zhz)
黄海学院重点科研项目(2020RW02)。
关键词
箱型主梁
动态随机重组
粒子群算法
轻量化
罚函数
Box Girder
Dynamic Random Regroup
Particle Swarm Algorithm
Lightweight
Penalty Function
